Chandana Withana

Enhanced Apriori Algorithm model in course suggestion system

Association Rule Mining (ARM) includes one of the most popular algorithms called Apriori Algorithm (AA) in it. AA has some limitations and areas for improvement related to the execution time consumption. The extended AA has some improvements based on the existing algorithm AA. The main findings indicate that not much has been done in an educational data mining considering the volume of data that is observed. Therefore, it is intended to evaluate to what extent an ARM can be utilized in an educational data mining context. Specifically, this paper develops an extended AA mining algorithm and applies it to the higher education system, focusing on the students course planning system. This project focuses on introducing AA and ARM, implementing the enhanced features of the AA in educational data, develop the course recommender model. Finally, it evaluates the enhanced algorithm compared to the existing AA to help build a course suggestion system for students.

Distributed Denial of Service attack on cloud: Detection and prevention

Cloud computing is a distributive and scalable computing architecture. It provides sharing of data and other resources which are accessible from any part of the world for a very low cost. However, Security is one major concern for such computing environment. Distributed Denial of Service (DDoS) is an attack that consumes all the cloud resources may have making it unavailable to other general users. This paper identifies characteristics of DDoS attack and provides an Intrusion Detection System (IDS) tool based on Snort to detect DDoS. The proposed tool will alert the network administrator regarding any attack for any possible resources and the nature of the attack. Also, it suspends the attacker for some time to allow the network admin to implement a fall back plan. As Snort is an open source system, modifying different parameters of the system showed a significant aid in not only detection of DDoS, but also reduction the time for the down time of the network. The proposed tool helps minimize the effect of DDoS by detecting the attack at very early stage and by altering with various parameters which facilitates easy diagnose of the problem.

Analyzing sensitivity of code execution efficiency to software and hardware parameters

Groups of code basic blocks have been employed in the past as software-related input parameters of processor models. In this paper, we use different block group variations to study their relationship to the execution efficiency of a superscalar processor. For this purpose, we utilized frequencies of blocks to represent dynamic nature of 10 benchmark programs. We were able to get a reduced set of hardware and software parameters by employing correlation coefficients and principal component analysis.